aboutsummaryrefslogtreecommitdiffstats
path: root/firmware
Commit message (Collapse)AuthorAgeFilesLines
* Merge branch 'master' into usrp1Josh Blum2010-08-177-29/+326
|\
| * usrp2: don't forward dbsm errorsNick Foster2010-08-161-4/+4
| |
| * IRQ-based SPI works. Don't try to do multiple transactions without waiting ↵Nick Foster2010-08-163-18/+20
| | | | | | | | | | | | for results first. In fact, don't try to do an I2C transaction while an SPI transaction is pending.
| * syntaxNick Foster2010-08-162-2/+3
| |
| * first stab at irq'ed spiNick Foster2010-08-163-6/+72
| |
| * Removed NACK checking so things don't barf when daughterboards aren't connected.Nick Foster2010-08-161-6/+7
| | | | | | | | When we have real error handling in the DUDE/BRO protocol we'll add this back in.
| * Fix warnings related to const-ness and volatile-ness.Nick Foster2010-08-162-7/+7
| |
| * Bring in changes since last week.Nick Foster2010-08-101-18/+10
| |
| * Interrupt-driven I2C. txrx_uhd uses async i2c for comms.Nick Foster2010-08-104-24/+259
| |
* | usrp1: Add SPI transaction command to FX2 firmwareThomas Tsou2010-08-134-3/+116
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The existing SPI read command is set for half-duplex 8-bit or 16-bit writes followed by an equivalent sized read. This patch adds a new command, VRQ_SPI_TRANSACT, which performs a full-duplex transaction up to 4-bytes. Since the data buffer of the USB control transfer is not available for outbound data on IN requests, the 4 bytes of write data are sent through the request parameters. Enables are sent in the previsouly unused high byte of the length parameter. The USB field mappings are shown below. Only rising edge and MSB operation is supported. Field (8-bit) Description ===== ===== bmRequestType 0x80 (USB Device IN request) bRequest 0x83 (VRQ_SPI_TRANSACT) wValueH OUT data(0) wValueL OUT data(1) wIndexH OUT data(2) wIndexL OUT data(3) wLengthH SPI enables wLengthL Number of bytes to transfer (1-4) EP0BUF IN data
* | usrp1: Add FX2 firmware filesThomas Tsou2010-08-13107-0/+11421
|/ | | | | These firmware files for the usrp1 are imported from GNURadio.
* usrp2: added fpga compat number, renamed firmware proto version to compatJosh Blum2010-08-091-5/+6
|
* usrp2: reorganized firmware directoryJosh Blum2010-07-3085-911/+278
|
* usrp2: added tx policy register and changed fw to not handle error in buffer ↵Josh Blum2010-07-291-1/+1
| | | | pool
* usrp2: installed a simple dbsm_rx_inspector, it wont work without it...Josh Blum2010-07-161-4/+7
|
* usrp2: remove mb intervention on outgoing packetsJosh Blum2010-07-151-28/+6
|
* usrp2: forward errors in the dbsm to communicate them up to the hostJosh Blum2010-07-121-2/+4
|
* usrp2: Added a peek64 to read pairs of 32 bit numbers such as time64Josh Blum2010-07-051-0/+8
| | | | also added a templated host to/from network conversion in byteswap.hpp (didnt use it though)
* usrp2: increased transport buffer minimum size, and added warningJosh Blum2010-07-051-4/+4
| | | | | | added more notes on buffer size to the manual pulled in some firmware fixes from the mimo work, just to have them in here
* usrp2 mb: replace hard coded constants with macros for rx offsetJosh Blum2010-06-181-2/+2
|
* Moved usrp2 eeprom addr read/write to host control over i2c/eeprom interface ↵Josh Blum2010-05-111-20/+3
| | | | | | | | | | calls. No longer part of the dude/bro protocol. Simplified the mac and ip addr read write code in mboard impl. Modified the mac addr type to take byte_vector_t from serial.hpp types.
* Added a place for serdes control on the host.Josh Blum2010-05-102-20/+20
| | | | | Fix bug in codec control. Comment out some clock control in fw code.
* Merge branch 'shrinkfw' into usrp2Josh Blum2010-05-1010-467/+4
|\
| * Moved some misc setting registers into host.Josh Blum2010-04-281-10/+0
| |
| * removed some unused things like gpio from microblaze codeJosh Blum2010-04-287-312/+4
| |
| * Added reload flag to the stream cmd.Josh Blum2010-04-285-147/+2
| | | | | | | | | | This reloads the last command to handle continuous streaming in hardware. Moved rx control register setup and stream command issuing to the host.
* | added icmp echo reply to usrp2Josh Blum2010-05-031-1/+18
|/
* Merge branch 'eeprom' of git@ettus.sourcerepo.com:ettus/uhdprivJosh Blum2010-04-265-127/+0
|\
| * Moved reading the eeprom (dboard ids) onto the host.Josh Blum2010-04-235-127/+0
| | | | | | | | | | Created a eeprom parser for the format in the dboard. Removed the support from the microblaze code.
* | renamed the firmware main image to txrx_uhd.binJosh Blum2010-04-252-2/+2
|/
* toggle led a for continuous streamingJosh Blum2010-04-221-0/+6
|
* Added protocol version number to usrp2 common header and data struct.Josh Blum2010-04-211-5/+29
|
* pulled aux dac and adc support from microblaze, its in the host nowJosh Blum2010-04-197-280/+0
|
* removed unused firmware codeJosh Blum2010-04-133-370/+0
|
* Moved ad9777 control 100% on to the hostJosh Blum2010-04-135-168/+0
|
* Reworked the spi part of the dboard interface.Josh Blum2010-04-071-18/+7
| | | | | | It turns out to be more complicated. The integer type is better for holding the bits. The edges can be different, so the spi config hold 2 edge setting for mosi and miso.
* added 16 bit peek and poke, 16 bit register defs for gpios and atrsJosh Blum2010-04-051-3/+27
|
* extended stream cmd with mode enum, and extended fragment flags in metadataJosh Blum2010-04-031-2/+2
|
* added spi slaves to regs, use std copy for buffsJosh Blum2010-03-311-1/+1
|
* Moved dsp (rx and tx), time config, and clock config (mostly) into the host.Josh Blum2010-03-311-116/+0
|
* Refactor ATR part of dboard interface (and some constants).Josh Blum2010-03-311-49/+24
| | | | | | Added peek and poke to the dude/bro protocol. Started moving more control code through peek and poke. Added usrp_regs.hpp to be like memory map for slave perifs.
* removed that fw sets seq stuff, not applicableJosh Blum2010-03-261-16/+1
|
* insert correct control word and vrt len into the bufferJosh Blum2010-03-261-19/+15
|
* Overhaullllllled the way we do streaming. There is an odd bug whereJosh Blum2010-03-252-81/+96
| | | | | | | | | a zero length command (now, no chain) used to stop the streaming. Now it seems to do the reverse... must investigate. Made all clock configuration into enums. The strings were painful and there cant be that many variations that enums cant cover them. The enums will make more sense to developers than mystery strings.
* added easy way to compile firmware for debug modeJosh Blum2010-03-241-7/+9
|
* Added recovery app to use raw socket to burn known ip addr.Josh Blum2010-03-164-36/+32
| | | | Fixed up fw so the device can reply after new addrs are set.
* Added ability to set the subdevices in use for rx and tx dboards.Josh Blum2010-03-162-4/+11
| | | | | This is used to calculate and set the ddc and duc muxes. Also, minor fix for burning addrs (wrong pointer....)
* Ability to burn mac addr and ip addr to usrp2 (over ip/udp for now).Josh Blum2010-03-154-36/+96
| | | | Added firmware support and usrp2 burner host app.
* Added simple device to handle wrapping general properties up into simple api.Josh Blum2010-03-101-0/+10
| | | | | Added setting time capability to the usrp2 impl. Messing with props and time specs...
* Moved timeouts into the udp transports.Josh Blum2010-03-043-18/+11
| | | | | | Simplified the fast path checking in the fw, but it turns out this was not the issue. Fixed some bad bit operations with the 16sc words (dont forget sign extension). Added some more documentation to the headers....
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-28 12:57:12 +0000